CN1902949A - Method and system of double engines sharing memory - Google Patents

Method and system of double engines sharing memory Download PDF

Info

Publication number
CN1902949A
CN1902949A CNA2004800398507A CN200480039850A CN1902949A CN 1902949 A CN1902949 A CN 1902949A CN A2004800398507 A CNA2004800398507 A CN A2004800398507A CN 200480039850 A CN200480039850 A CN 200480039850A CN 1902949 A CN1902949 A CN 1902949A
Authority
CN
China
Prior art keywords
veneer
message
communication
master control
control borad
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CNA2004800398507A
Other languages
Chinese (zh)
Other versions
CN100576936C (en
Inventor
寻劲由
吴思欣
徐继军
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
UTStarcom Telecom Co Ltd
Original Assignee
UTStarcom China Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by UTStarcom China Co Ltd filed Critical UTStarcom China Co Ltd
Publication of CN1902949A publication Critical patent/CN1902949A/en
Application granted granted Critical
Publication of CN100576936C publication Critical patent/CN100576936C/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04Q—SELECTING
    • H04Q1/00—Details of selecting apparatus or arrangements
    • H04Q1/18—Electrical details
    • H04Q1/28—Current-supply circuits or arrangements for selection equipment at exchanges

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Hardware Redundancy (AREA)
  • Small-Scale Networks (AREA)

Abstract

The present invention discloses a communication mechanism of the main control boards which is separated from specific service and constructed on normal CompactPCI bus with the peripheral single boards, which particularly satisfies the communication mechanism of high performance and high reliability required in the communication devices, the communication mechanism completes the switch between the main control boards and peripheral single boards, and between the peripheral single boards, in which the core module is high speed sharing memory engine, in the main control boards which are mutually standby, there have respectively a high speed sharing memory engine, and both the high speed sharing memory engine are hot standby mutually. when one engine fails to work or the main control board which the engine resides fails to work, another engine takes the work without gap, and any data information won't lost.

Description

Method and system of double engines sharing memory
Shared drive communication means with double engines and systems technology field
The present invention relates to the shared drive communication means with double engines in communication system, including GSM, fixed communications and data communication system and system.Background technology
In general, this kind of Working level system of such as GSM, fixed communications and data communication system is different from individual application system, and they belong to the system of requirement high reliability.And this kind of system is more more complicated than general system, there are many service sub-systems in a system, and interknited between each service sub-system, there is substantial amounts of business datum to need to intercourse.Each subsystem may be operated on different physical entities again, such as between different machine frames, or the different veneer of identical machine frame.
In the case of running different service sub-systems on the different veneers of same machine frame, each service sub-system is usually present the exchange of data-message from each other.For example,
In UTStarcom SGW, signalling system No.7 management module, system HA modules are run on master control borad, the MTP2 modules of certain block veneer operation wherein run MTP3 modules on certain block veneer, Sigtran modules are run on other veneer.There is the exchange of management message and the mutual transmission of protocol data bag between these modules.
The interaction of data-message between different business is formed with different requirements to the conveyer of bottom.For example, time delay 4 blunt sensitivity of some subsystems to message, such as in the No.7 nation-wide Network Central Shanxi Plain, one message enters SGW from traditional No.7 nation-wide Network, by the place of gateway, reason, from the output of IP networks, requirement of the time delay no more than national Specification during this.Some subsystems are not very sensitive to the time delay of message, but it is very sensitive to whether message reliably reaches, whether message sequentially reaches, for example in the No.7 nation-wide Network Central Shanxi Plain, signaling management module on master control borad sends out an administration order to veneer, if this order lost, being likely to result in signaling link can not use.The possible message data block of subsystem also is very big, but is not intended to carry out burst to message again, thus there is special requirement to maximum message length. Because in complicated communication system, the communication between subsystems is inevitable, and different subsystems has different requirements, it is then desired to the communication means between a kind of different desired subsystems that disclosure satisfy that subsystems.
It is briefly described below currently used for the two kinds of existing communication mechanism solved the above problems.
1. the communication mechanism based on Ethernet.
Each veneer has operation ICP/IP protocol stack on minimum of one Ethernet interface, veneer.Now, if each veneer directly goes out Ethernet interface, then mean to need to be further provided with corresponding interchanger outside system；And if by the backboard ether bus of PIGMG2.16 standards, then meaning to need in system to add a veneer for serving as function of exchange in addition.No matter any mode, communication between each subsystem is all framed on ICP/IP protocol stack, this means that, if taking UDP communications to there is the problem of Xiao Xi Lost are lost, order is reached, and if taking TCP communication there is manages all TCP connections and the problem of system HA is relatively difficult to realize.For example, one veneer is by TCP and master control board communications, and when master control borad finds that mistake is switched to standby master control borad, the connection of the TCP will be cut off, connection between veneer and master control borad needs to re-establish, and the message data of such veneer and master control borad will lose.In addition, the problem of message delay all there may be based on TCP or UDP, because message all has to pass through two complete ICP/IP protocol stacks from a subsystem to another subsystem, and the performance of protocol stack turns into a problem that must be solved.
2. the communication mechanism based on serial ports.
For this communication mechanism, if between each veneer, veneer is with being required for communication between master control borad, then the Serial Port Line that these plates must have physics two-by-two is connected.If not designing these lines from backboard, and connected from panel or back plate, will cause that system is very chaotic from structure；And if connecting these Serial Port Lines from backboard, then backboard will be nonstandard.Also, if using the veneer for having mat woven of fine bamboo strips tripartite, it will have no idea to be in communication with each other third-party veneer and the veneer of oneself.Further, the bandwidth of serial ports will be the bottleneck that communicates between subsystems.
In view of the particular/special requirement of above-mentioned exploitation next generation communication system, both above communication mechanism respectively has shortcoming, the requirement of the communication system based on standard platform can not be all met.
In addition, the VME buses used in existing communication mechanism, the bus standard proposed as the eighties, there is certain limitation, such as its theoretical bandwidth is 80Mbps ,-as it is real 40Mbps can only be arrived in the application of border, and cPCI can reach 264Mbps in theory.The producer for supporting VME buses simultaneously be not it is a lot, the board of other so quick integrated manufacturers and quickly develop new product will be extremely difficult.The content of the invention
It is therefore an object of the present invention in order to overcome above mentioned problem present in existing communication mechanism there is provided a kind of shared drive communication means with double engines and system, to realize the high-performance and high reliability of shared drive.
The present invention discloses a kind of master control borad and the communication mechanism of peripheral single board independently of specific transactions being framed in the CompactPCI buses of standard, especially meet the highly reliable communication mechanism of the high-performance required in a communications device, this communication mechanism completes data master control borad in a communications device and peripheral single board, and the exchange between peripheral single board, nucleus module therein is the shared drive engine of high speed, in the master control borad mutually backed up, respectively there is the shared drive engine of a high speed, the two mutual Hot Spares of shared drive engine, one engine breaks down, or the master control borad at place breaks down, what another engine was seamless takes over work, there is no the loss of any data-message.
According to an aspect of the present invention there is provided a kind of communication system, include the main control board and the veneer of at least one operation subservice module of operation main business module, it is characterised in that：The veneer includes being divided into tetrameric physical memory, wherein, Part I storage veneer sends a message to the message header of other plates, Part II storage veneer receives the message header of the message of other plates, the part of the mat woven of fine bamboo strips three is to deposit the message body that veneer sends a message to other plates, and Part IV storage veneer receives the message body of the message of other plates；And the master control borad includes a Virtual and intends internal memory, the physical memory for mapping the veneer.
According to another aspect of the present invention, a kind of service board used in above-mentioned communication system is provided, it is characterized in that the veneer includes being divided into tetrameric physical memory, wherein, mat woven of fine bamboo strips part storage veneer sends a message to the message header of other plates, Part II storage veneer receives the message header of the message of other plates, and the part of the mat woven of fine bamboo strips three is to deposit the message body that veneer sends a message to other plates, and Part IV storage veneer receives the message body of the message of other plates.
There is provided a kind of master control used in above-mentioned communication system according to another aspect of the present invention Plate, it is characterised in that the master control borad includes a Virtual and intends internal memory, for the physical memory of the service board used in mapping communication system.
There is provided a kind of shared drive communication means with double engines based on cPCI buses in accordance with a further aspect of the present invention, it is characterised in that including step：A) shared drive of mapping veneer is to master control borad virtual memory；B) poll veneer shared drive header fields, until obtaining an information header；C) according to acquired information header, Target Board is determined；D) position of the imformosome in shared drive, the shared drive information of Target Board are obtained；And e) start direct memory read-write, completion message transmission.
In addition, also provide computer program product of the record at least one computer-readable Jie Skin, including when computer in use, making the functional descriptions material for the step of computer performs above-mentioned shared drive communication means with double engines based on cPCI buses.
According to the above method and system of the present invention, the problem of bandwidth is solved using the cPCI buses of 64 66M frequencies, mechanism based on shared bus ensure that the order of message is reached, and the communication means based on internal memory direct read/write solves the delay requirement to message transmission.Brief description of the drawings
Fig. 1 shows to run the situation of different service sub-systems on the different veneers of same machine frame；
Fig. 2 shows the physical memory distribution of each veneer and the virtual memory mappings advocated peace from master control borad；
Fig. 3 shows the block diagram of the master control borad of communication system according to an embodiment of the invention；
Fig. 4 shows to carry out the flow chart of master control borad during main and standby boards switching in communication system according to an embodiment of the invention；
Fig. 5 shows the flow chart of the new veneer insertion process in communication system according to an embodiment of the invention；
Fig. 6 shows the flow chart of the veneer withdrawal process in communication system according to an embodiment of the invention；And
Fig. 7 shows the high reliability with double engines based on cPCI buses according to the present invention in operation Shared drive communication means when master control borad on performed basic handling flow.Embodiment
It is an object of the invention to provide a kind of communication system and method for the high reliability of the shared drive of measured cPCI buses, and the communication means is independently of any specific business.For example：It is can apply to according to the communication means of the present invention in SGW, the reliable message of high speed of each protocol layer is transmitted, the communication of submodule that can also be applied in Tandem Gateway, be distributed on each veneer, apply also in call server, the interaction of call information between each service sub-system, mirror image etc..
Below the present invention is illustrated exemplified by running different service sub-systems on the different veneers in same machine frame.
Here first it should be noted that, following description is merely exemplary matter, the subsystem that the present invention is not limited only to run different business is located at the situation of same machine frame, and is operated in including each subsystem on different physical entities, such as the situation of different machine frame.
Fig. 1 shows to run the situation of different service sub-systems on the different veneers of same machine frame, wherein, there is the exchange of data-message from each other in each service sub-system.As shown in Fig. 1, in main control board and the operation main business module from master control borad, and each subservice module is run on veneer 1 to single power N.For example, on UTStarcom SGW, signalling system No.7 management module, system HA modules are run on master control borad, certain block veneer 1 is to running MTP2 modules on veneer N wherein, MTP3 modules are run on certain block veneer 1 to veneer N, Sigtran modules are run on other veneer 1 to veneer N.There is the exchange of management message and the mutual transmission of protocol data bag between these modules.The interaction of all these modules is all completed by the communication means for the shared drive of the invention based on high reliability with double engines being described more fully below.
Herein it should also be noted that, the bus standard proposed as the eighties, VME buses have certain limitation, and such as its theoretical bandwidth is 80Mbps, 40Mbps can only be arrived in general practical application, and cPCI can reach 264Mbps in theory.In addition, the producer for supporting VME buses is not a lot, not as cPCI buses, there are many mat woven of fine bamboo strips tripartite manufacturers to provide the cPCI plates of various functions.Therefore, preferably, the present invention uses cPCI Bus is used as each plate（Including master control borad and veneer）Between connection bus, be so easy to the board of other quick integrated manufacturers, so as to quickly develop new product.
According to the communication system of the present invention, each veneer in above-mentioned machine frame and advocate peace and distribute the physical memory of one section of suitable size according to specific demand from master control borad, the application heap on each veneer is divided into four most of, the physical memory distribution for each veneer being specifically shown in shown in Fig. 2 and Fig. 3 and the virtual memory mappings advocated peace from master control borad simultaneously.
As shown in Fig. 2 distributing one section of physical memory PM-DAT on each piece of veneer and being divided into four parts.Wherein, Part I storage veneer sends a message to the message header P-M-Q of other plates, Part II storage veneer receives the message header M-P-Q of the message of other plates, and Part III is to deposit veneer and send a message to the message body P-M-D of other plates (to be designated as DATA parts in figure）, the message body M-P-D of the Part IV storage veneer message that receives other plates (is designated as DATA parts in figure）.
The each section of aforementioned four part is divided into three segments again.For example：There are P-M-Q-H, P-M-Q-N. and P-M-Q-L in the first portion, mean respectively, veneer sends a message to other plates P-M-Q-H with limit priority, veneer sends a message to other plates P-M-Q-N with normal priority, and veneer sends a message to other plates P-M-Q-L with low priority.
Meanwhile, corresponding Virtual is mapped on master control borad and intends internal memory VM-DAT, the above-mentioned physical memory PM-D AT with each veneer are corresponding respectively.
In addition, as shown in Figure 3, also include global control block SME-CB, communication engines higher management module SMP, communication engines management module SMM, HA management module HAM, real-time data base-message route mark SM-MRT and unshowned hot-swappable detection module HSD on master control borad.These modules are mirror images in main control board and from master control borad.
Global control block SME-CB saves corresponding relations of the physical memory PM-D AT with virtual memory VM-DAT on master control borad of each veneer.
Communication engines higher management module SMP is a priority very high module, and it ceaselessly inquires about the Part I of the shared drive of each veneer according to priority policy.If communication engines higher management module SMP has found that certain block veneer has message to be sent to other plates, the message, and the purpose subsystem being sent on purpose plate are taken out from the internal memory of this block veneer immediately.
Communication engines management module SMM has heartbeat mechanism with each veneer, manages each veneer State.Communication engines management module SMM also has message interface with the HA management modules HAM of system, it is possible to stop and start communication engines higher management module SMP.
In addition, it is just blunt according to communication system of the invention, a system-level HA management modules HAM is run on each master control borad.HA management modules HAM manages the health status of whole system, and active-standby switching service is provided for whole system.Once mistake occurs in main control board, it notifies communication engines management module SMM, communication engines management module SMM then to stop current communication engines higher management module SMP immediately immediately.So as to which HA management modules HAM will be changed into current main control board from master control borad, and handle is changed into main control board from master control borad, and notify communication engines management module SMM.Start communication engines higher management module SMP immediately after communication engines management module SMM receives the notice.So, second communication engines starts immediately.Because shared drive is stored on veneer, so the loss without any message data, the very smooth switching of two communication engines comes.
A real-time data base-message route mark SM-MRT is additionally provided with master control borad.Can more preferable dispatch messages according to the message route mark SM-MRT, communication engines higher management module SMP.Because if the submodule on a veneer will send a message to another veneer, this 4 blunt benefit reason for communication engines, it can directly transmit this message to destination according to the groove number of purpose veneer.
But, sometimes the submodule on veneer does not know that the groove number of purpose veneer, or target single board carry out redundancy backup using standby usage or other modes, and the ID of at this moment submodule meeting application target module is addressed.So, communication engines higher management module SMP can just translate into module I D according to message route mark SM-MRT the purpose plate just in processing business.
When purpose plate is switched over, outside management module will coordinate with communication engines higher management module SMP, complete handoff procedure.Fig. 4 shows the flow performed by master control borad during the switching of standby usage veneer.
First, as shown in figure 4, in step SP41, stopping the business on main veneer.After primary veneer, which obtains oneself, will be switched to standby notice, the local search task on primary veneer will stop inquiring about.At this moment, there is Slow in the shared drive of this veneer in the message on the service channel.
Then, Bu Sudden SP42, send the message not yet distributed on the main veneer.Then, In step SP43, the message not received on the main veneer is sent to standby board.Then, in step SP44, the switching of standby usage veneer is completed, the routing iinformation of main and standby relation is updated.The message " fetching " that Slow is deposited on old plate is sent to the new veneer for handling the business by communication engines higher management module SMP.Meanwhile, message route mark SM-MRT is updated, and service message hereafter will be sent directly to new service board.In the process, if the shared drive of old plate can not be accessed（Corresponding to due to the extremely caused uncontrolled switching of old plate software and hardware）, then " fetching " process is skipped, the Xiao Xi Lost for being already sent to old plate lose.
Afterwards, in step SP45, enable the information on new main veneer and receive and send service.New main veneer is started a query at and processing business.
Preferably, it is route in message during target is accessed and mounts the interpretative function of extension so as to realize the function of load sharing.
In addition, having using cPCI hot-swappable（Hot Swap) characteristic, veneer in systems hot-swappable is also supported according to the communication system of the present invention.Fig. 5 and Fig. 6 illustrate new veneer insertion process and veneer withdrawal process respectively.
As shown in figure 5, in step SP51, when inserting new veneer in machine frame, the hot-swappable detection module HSD on master control borad detects the event and board type is judged.When this veneer supports shared drive engine, Bu Sudden SP52, hot-swappable detection module HSD is configured to the pci interface chip of the veneer, for the shared drive mapping PCI addresses on the veneer.Then, in step SP53, relevant parameter is notified communication engines management module SMM by hot-swappable detection module HSD.
Afterwards, in step SP54, communication engines management module SMM starts the heartbeat detection of the veneer.After heartbeat detection success, in step SP55, corresponding relations of the physical memory PM-DAT with virtual memory VM-DAT on master control borad of the veneer preserved in global control block SME-CB is updated, and confirms to control communication engines higher management module SMP after the single board activated in step SP56 for the single board starting communication service.So as to complete the insertion process of new veneer.
For the withdrawal process of veneer, as shown in fig. 6, in step SP60, when under the hot-swappable detection module HSD plug spanners for detecting certain veneer are pulled, hot-swappable detection module HSD is by the message informing plate driver in step chaff SP61.If also, having been opened on this veneer Shared drive communication service is moved, then the hot-swappable detection module HSD of Bu Sudden SP62 notify communication engines management module SMM immediately.
Afterwards, in step SP63, communication engines management module SMM controls communication engines higher management module SMP stops the communication service of the veneer.Then, Bu Sudden SP64, communication engines management module SMM stop nullifying the plate to global control block SME-CB in the plate heartbeat detection, and Bu Sudden SP65.
Complete after aforesaid operations, notify hot-swappable detection module HSD to pull out plate processing and terminate in step SP66, the veneer can be safeguarded with normally being extracted in step Sudden SP67.
The shared drive communication system and the processing procedure of hot-swappable service board within the system of the high reliability with double engines based on cPCI buses according to the present invention are described in detail above in association with accompanying drawing.
The shared drive communication means of the high reliability with double engines based on cPCI buses according to the present invention is described in detail below in conjunction with accompanying drawing.
Fig. 7 shows the basic handling flow performed on master control borad in shared drive communication means of the operation according to the high reliability with double engines based on cPCI buses of the present invention.
As shown in fig. 7, in step SP71, master control borad maps the shared drive of each veneer to this plate address space.Then, in step SP72, record veneer shared drive information and function-groove number mapping.
After above-mentioned initial work is completed, in step SP73, master control borad starts poll veneer shared drive header fields, until obtaining an information header.
Then, judge to whether there is purpose groove number in the information header in step SP74.If there is purpose groove number in information header, the information header is sent to the veneer of groove number specified by purpose groove number in Bu Sudden SP75.Afterwards, handling process proceeds to step SP78.
If judging purpose groove number is not present in information header in step SP74, the information header is sent to specified function veneer in step SP76.Then, in step SP77 by function, main and standby relation, meet the relation of sharing and translate purpose groove number.
Then, in Bu Sudden SP78, master control borad obtains position of the imformosome in shared drive, the shared drive information of Target Board, and starts direct memory read-write according in acquired above- mentioned information Bu Sudden SP79.So as to complete the transmission of an information in step SP80. After the transmission that an information is completed in step SP80, handling process return to step SP73, master control borad continues poll veneer shared drive header fields, to carry out the transmission of next information.
In addition, being provided with a set of API on veneer to specific business module, easily to use the communication means according to the present invention, each business module sends a message to the message that other plates and receiving are sent from other plates using this set of API.There is a local enquiry module on veneer, the shared drive block of Part II is ceaselessly inquired about according to predetermined priority policy：M-P-Q, if message is reached, then notifies specific business module received message immediately.
It can be seen that from described in detail above, according to the shared drive communication system and method for the high reliability with double engines based on cPCI buses of the present invention, its M-P-Q/P-M-Q mechanism make it that direct memory read-write is possibly realized when messaging path is " veneer-veneer ".And the CPU decision message route of master control borad, do not participate in message copying.Because the traffic of " veneer-veneer " in system accounts for 95 more than % of the whole traffic, so that, data communication can be realized with high performance according to the shared drive communication system and method for the high reliability with double engines based on cPCI buses of the present invention.
Further, since shared drive is present on veneer, rather than on master control borad, therefore, when master control borad switches, the information in shared drive will not lose.So as to for the complete, unique of message is completely secured and most basic guarantee condition is provided in proper order.
In addition, SM-MRT mechanism then causes the peripheral single board on cPCI to realize active-standby switch when being communicated with other veneers using shared drive.Hot-swappable detection and heartbeat detection mechanism then ensure it is most fast detect single board default, so as to ensure the Message Time Delay in veneer handoff procedure.
" fetching " process ensure that the complete of in veneer handoff procedure message, and " with double engines " then ensures that shared drive communication remains to be normally carried out in master control borad fault.
Thus, it is possible to find out, according to the shared drive communication system and method for the high reliability with double engines based on cPCI buses of the present invention, it is possible to achieve the high-performance and high reliability of whole communication system.
The present invention is illustrated while in accordance with preferred embodiment above, but the present invention can be further changed in the spirit and scope of present disclosure.So as to, the application covered by its general principles the present invention any modification, using or modification.In addition, it is intended to cover from this public affairs Known within open that content sets out technically relevant with the present invention and restriction in appended claims or commercial practice.

Claims (1)

  1. Claim
    1. a kind of communication system, includes the main control board and the veneer of at least one operation subservice module of operation main business module, it is characterised in that：
    The veneer includes being divided into tetrameric physical memory (PM-D AT), wherein, Part I storage veneer sends a message to the message header (P-M-Q) of other plates, Part II storage veneer receives the message header (M-P-Q) of the message of other plates, Part III is to deposit the message body (P-M-D) that veneer sends a message to other plates, and Part IV storage veneer receives the message body (M-P-D) of the message of other plates;And
    The master control borad includes a Virtual and intends internal memory (VM-DAT), the physical memory (PM-DAT) for mapping the veneer0
    2. the communication system as described in claim 1, it is characterised in that the communication system is based on cPCI buses.
    3. the communication system as described in claim 2, it is characterised in that each section for being divided into tetrameric physical memory (PM-D AT) is divided into three segments of high, medium and low priority again.
    4. the communication system as described in claims 1 to 3 is any, it is characterised in that the communication system is also included with main control board symmetrical configuration from master control borad.
    5. the communication system as described in claim 4, it is characterised in that the main control board and include respectively from master control borad mirror image：
    Global control block (SME-CB), the corresponding relation of physical memory (PM-D AT) and virtual memory (VM-D AT) on master control borad for preserving veneer；And
    Communication engines higher management module (SMP), according to the Part I of the shared drive of priority policy poll veneer, if it find that certain block veneer has message to be sent to other plates, then take out the message, and the purpose subsystem being sent on purpose plate from the internal memory of this block veneer immediately.
    6. the communication system as described in claim 5, it is characterised in that the main control board and include respectively from master control borad mirror image：
    , there is heartbeat mechanism in communication engines management module (SMM), the state for management single board with veneer；And
    HA management modules (HAM), the health status for managing whole system is whole system System provides active-standby switching service；And
    Communication engines management module (SMM) also has message interface with the HA management modules (HAM) of system, it is possible to stop and start communication engines higher management module (SMP).
    7. the communication system as described in claim 6, it is characterized in that the main control board and from master control borad mirror image respectively include message route mark (SM-MRT), and communication engines higher management module (SMP) according to the message route mark (SM-MRT) dispatch messages.
    8. the service board used in a kind of communication system as described in as any such as claim 1-7, it is characterized in that the veneer includes being divided into tetrameric physical memory (PM-DAT), wherein, Part I storage veneer sends a message to the message header (P-M-Q) of other plates, Part II storage veneer receives the message header (M-P-Q) of the message of other plates, Part III is to deposit the message body (P-M-D) that veneer sends a message to other plates, and Part IV storage veneer receives the message body (M-P-D) of the message of other plates.
    9. the service board as described in claim 8, it is characterised in that each section for being divided into tetrameric physical memory (PM-D AT) is divided into three segments of high, medium and low priority again.
    10. the master control borad used in a kind of communication system as described in as any such as claim 1-7, it is characterized in that the master control borad, which includes a Virtual, intends internal memory (VM-DAT), for the physical memory (PM-DAT) of the service board used in mapping communication system.
    11. master control borad as claimed in claim 10, it is characterised in that physical memory (PM-DAT) is divided into four parts, wherein, Part I storage veneer sends a message to the message header of other plates
    (P-M-Q), Part II storage veneer receives the message header (M-P-Q) of the message of other plates, the part of the mat woven of fine bamboo strips three is to deposit the message body (P-M-D) that veneer sends a message to other plates, and Part IV storage veneer receives the message body (M-P-D) of the message of other plates.
    12. master control borad as claimed in claim 11, it is characterised in that each section for being divided into tetrameric physical memory (PM-DAT) is divided into three segments of high, medium and low priority again.
    13. the master control borad as described in claim 10 to 12 is any, it is characterised in that the master control borad also includes：
    Global control block (SME-CB), the corresponding relation of physical memory (PM-DAT) and virtual memory (VM-D AT) on master control borad for preserving veneer；And
    Communication engines higher management module (SMP), according to the shared of priority policy poll veneer The Part I of internal memory, if it find that certain block veneer has message to be sent to other plates, then takes out the message, and the purpose subsystem being sent on purpose plate from the internal memory of this block veneer immediately.
    14. the master control borad as described in claim 10 to 12 is any, it is characterised in that the master control borad also includes：
    , there is heartbeat mechanism in communication engines management module (SMM), the state for management single board with veneer；And
    HA management modules (HAM), the health status for managing whole system provides this from the service of switching for whole system；And
    Communication engines management module (SMM) also has message interface with the HA management modules (HAM) of system, it is possible to stop and start communication engines higher management module (SMP).
    15. the master control borad as described in claim 10 to 12 is any, it is characterized in that the master control borad also includes message route mark (SM-MRT), and communication engines higher management module (SMP) is according to message route mark (SM-MRT) dispatch messages.
    16. a kind of shared drive communication means with double engines based on cPCI buses, it is characterised in that including step：
    A) shared drive of mapping veneer intends internal memory to master control borad Virtual；
    B) poll veneer shared drive header fields, until obtaining an information header；
    C) according to acquired information header, Target Board is determined；
    D) position of the imformosome in shared drive, the shared drive information of Target Board are obtained；And
    E) direct memory read-write, completion message transmission are started.
    17. communication means as claimed in claim 16, it is characterised in that also include the step of record veneer shared drive information and function-groove number mapping before step b).
    18. communication means as claimed in claim 17, it is characterised in that step c) comprises the following steps：
    Judge to whether there is purpose groove number in the information header；
    If it is determined that there is purpose groove number in information header, then the information header is sent to the veneer of groove number specified by purpose groove number；And
    If it is determined that purpose groove number is not present in information header, then the information header is sent to specified Function veneer, and according to function, main and standby relation, meet the relation of sharing and translate purpose groove number.
    19. the communication means as described in claim 16 to 18 is any, it is characterised in that veneer is just blunt ceaselessly to inquire about its shared drive block according to predetermined priority policy, if message is reached, then notifies specific business module received message immediately.
    20. the communication means as described in claim 19, it is characterised in that when standby usage veneer switches, also comprise the following steps：
    Stop the business on main veneer；
    Send the message not yet distributed on the main veneer；
    The message not received on the main veneer is sent to standby board；
    Update the routing iinformation of main and standby relation；And
    Enable the information on new main veneer and receive and send service.
    21. the communication means as described in claim 19, it is characterised in that when inserting new veneer, also comprise the following steps：
    Hot-swappable detection module (HSD) on master control borad detects the insertion of new veneer and board type is judged；
    When this veneer supports shared drive engine, hot-swappable detection module (HSD) configures to the pci interface chip of the veneer, for the shared drive mapping PCI addresses on the veneer；Relevant parameter is notified communication engines management module (SMM) by hot-swappable detection module (HSD);
    Communication engines management module (SMM) starts the heartbeat detection of the veneer；
    Update global control block（SME-CB the physical memory of the veneer preserved in)
    (PM-DAT) with virtual memory on master control borad (VM-DAT) corresponding relation；And
    Confirm to control communication engines higher management module (SMP) to be the single board starting communication service after the single board activated.
    22. communication means as claimed in claim 19, it is characterised in that when extracting veneer, also Bao includes Xia Lie Bu Sudden：
    When under the plug spanner that hot-swappable detection module (HSD) detects certain veneer is pulled, by the message informing plate driver, if having been started up shared drive communication service on this veneer, communication engines management module (SMM) is notified immediately; Communication engines management module (SMM) control communication engines higher management module (SMP) stops the communication service of the veneer；
    Communication engines management module (SMM) stops the plate heartbeat detection, and nullifies the plate to global control block (SME-CB)；And
    Hot-swappable detection module (HSD) is notified to pull out plate processing and terminate.
    23. a kind of record the computer program product at least one computer-readable medium, including when computer in use, making the functional descriptions material of the method step described in any one in computer perform claim requirement 16-22.
CN200480039850A 2004-02-27 2004-02-27 Shared drive communication means with double engines and system Expired - Fee Related CN100576936C (en)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2004/000154 WO2005084040A1 (en) 2004-02-27 2004-02-27 A method and a system of double engines sharing memory

Publications (2)

Publication Number Publication Date
CN1902949A true CN1902949A (en) 2007-01-24
CN100576936C CN100576936C (en) 2009-12-30

Family

ID=34894643

Family Applications (1)

Application Number Title Priority Date Filing Date
CN200480039850A Expired - Fee Related CN100576936C (en) 2004-02-27 2004-02-27 Shared drive communication means with double engines and system

Country Status (2)

Country Link
CN (1) CN100576936C (en)
WO (1) WO2005084040A1 (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101068140B (en) * 2007-06-27 2010-06-16 中兴通讯股份有限公司 Apparatus and method for realizing primary/standby PCI equipment switchover
CN104669268B (en) * 2013-11-26 2016-08-03 中国科学院沈阳自动化研究所 A kind of redundancy underwater robot self-control system based on Hot Spare and method
CN105204977A (en) * 2014-06-30 2015-12-30 中兴通讯股份有限公司 System exception capturing method, main system, shadow system and intelligent equipment
CN113411237B (en) * 2021-08-18 2021-11-30 成都丰硕智能数字科技有限公司 Method, storage medium and system for detecting terminal state with low delay

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2352064A (en) * 1999-07-13 2001-01-17 Thomson Training & Simulation Multi-processor system with PCI backplane
CN1109416C (en) * 2000-04-25 2003-05-21 华为技术有限公司 Method and equipment for swapping active with standby switches
CN1197018C (en) * 2001-03-01 2005-04-13 中兴通讯股份有限公司 Device and method for implementing dual system slots
CN100336373C (en) * 2001-12-12 2007-09-05 中兴通讯股份有限公司 Remote file transmission method
KR100408047B1 (en) * 2001-12-24 2003-12-01 엘지전자 주식회사 Processor board dualizing apparatus and method for mobile communication system

Also Published As

Publication number Publication date
WO2005084040A1 (en) 2005-09-09
CN100576936C (en) 2009-12-30

Similar Documents

Publication Publication Date Title
CN102439903B (en) Method, device and system for realizing disaster-tolerant backup
CN108696569B (en) System and method for providing data replication in NVMe-oF Ethernet SSD
US6760859B1 (en) Fault tolerant local area network connectivity
US8619552B2 (en) Virtual router with a priority value per port
KR100831639B1 (en) Information processing apparatus, communication load decentralizing method, and communication system
CN101110776B (en) Backup method, backup device and backup system for data business
EP2091180A1 (en) Method, system and device for realizing switching between main and spare
TWI410084B (en) Intelligent failover in a load-balanced networking environment
CN102143046A (en) Load balancing method, equipment and system
US20210286747A1 (en) Systems and methods for supporting inter-chassis manageability of nvme over fabrics based systems
CN108259629A (en) The switching method and device of virtual IP address
US9807051B1 (en) Systems and methods for detecting and resolving split-controller or split-stack conditions in port-extended networks
CN112367257B (en) Route notification method and device
CN107766181A (en) A kind of dual controller storage High Availabitity subsystem based on PCIe non-transparent bridges
US11936762B2 (en) Processing protocol packet
CN113132159A (en) Storage cluster node fault processing method and device and storage system
JP5408361B2 (en) SWITCH DEVICE, INFORMATION PROCESSING DEVICE, SWITCH DEVICE CONTROL METHOD, AND PROGRAM
CN1902949A (en) Method and system of double engines sharing memory
EP2302517B1 (en) Storage system and method for use in a storage system
CN104243304A (en) Data processing method, device and system of locally-connected topological structure
CN100362484C (en) Method of multi-computer back-up
KR100429899B1 (en) Fault tolerant control apparatus and method by FUDCOM
WO2010028553A1 (en) Method, single board and system for port configuration management
CN101695029A (en) Data backup method, active-standby switching method and communication system as well as related equipment
CN115794769B (en) Method for managing high-availability database, electronic equipment and storage medium

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
ASS Succession or assignment of patent right

Owner name: UT STARCOM COMMUNICATION CO., LTD.

Free format text: FORMER OWNER: UT STARCOM (CHINA) CO., LTD.

Effective date: 20121225

C41 Transfer of patent application or patent right or utility model
COR Change of bibliographic data

Free format text: CORRECT: ADDRESS; FROM: 100000 DONGCHENG, BEIJING TO: 310053 HANGZHOU, ZHEJIANG PROVINCE

TR01 Transfer of patent right

Effective date of registration: 20121225

Address after: Hangzhou City, Zhejiang province Binjiang District 310053 Spring Road No. 1576 Leton Technology Park Building No. 3

Patentee after: UT Starcom Communication Co., Ltd.

Address before: 100000 Beijing city Dongcheng District East No. 1 Chang'an Avenue Oriental Plaza two east 10 storey office building

Patentee before: UT Sidakang (China) Co., Ltd.

CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20091230

Termination date: 20160227

CF01 Termination of patent right due to non-payment of annual fee